Application Of Satellite Image Classification Technology In Landslide Potential Analysis And Construction Of Risk Assessment Model

• Main Authors: Lin, Wei-Chung, 林洧全
• Other Authors: Chen, YIE-RUEY
• Format: Others
• Language: zh-TW
• Published: 2012
• Online Access: http://ndltd.ncl.edu.tw/handle/57963576190479868136

碩士 === 長榮大學 === 土地管理與開發學系碩士班 === 100 === In recent years due to technological and economic development the original plain area in Taiwan has been nearly fully developed, and human development has extended down to the hillside area. Human development coupled with the global impact of extreme weather, typhoons and heavy rains caused the hillside disaster. The scope and impact extent of the damage are more serious than ever before. For this reason, the establishment of a risk assessment model to assess landslide hazard event is essential to disaster prevention. In this study, the Genetic Adaptive Neural Network was implemented in high-resolution texture analysis techniques for the interpretation of satellite images of the sub-categories and to obtain surface information and hazard log data, and multivariate instability index to build landslide disaster latent sense of pattern. Ten landslide hazard potential factors are included: slope, geology, elevation, distance from the fault, distance from water, terrain roughness, slope roughness, effective accumulated rainfall, and developing situation. The potential degree of correlation between factors was tested through the correlation test. To study satellite images before and after typhoon Morakot, the image subtraction and normalized vegetation are used to calculate index for disaster information and logarithmic normal distribution, the probability of landslide disasters. A long interview with the local residents, review of the relevant literature, and setting out the conditions for the preservation of objects of the probability of victimization were carried out to calculate the spatiotemporal impact of the study area, vulnerable area to flooding and resistance level of disaster probability of exposure of the population. The regional risk map was plotted with the help of geographic information system and the landslide hazard risk assessment model. The results show that the landslide disaster predicted by this study is consistent with the information provided on the website of Soil and Water Conservation Bureau. From the risk map, the high-risk zone is concentrated in the junction of Taoyuan and Bora in Liouguei area. Therefore, the management authority should propose countermeasures for this area to achieve mitigation purposes.